Phidata
ActiveDescription
Phidata is a framework for building AI agents with memory, knowledge, and tool integration to make agents more capable and useful.
Key Features
- Production API — 50+ endpoints with SSE and WebSocket to build products on top
- 100+ pre-built tool integrations — Connect to external tools via pre-built toolkits
- Context Providers — Access live context from Slack, Drive, Wiki, MCP and more
- Human Approval — Pause runs for confirmation, tool-level admin approval
- Observability — Built-in OpenTelemetry tracing, run history, and audit logs
- Security — JWT-based RBAC with multi-user, multi-tenant isolation
Use Cases
💡 Build intelligent customer service agents with memory capabilities
💡 Integrate knowledge bases to build enterprise RAG applications
💡 Query data and generate reports through tool calls
💡 Deploy controllable agent platforms in your own cloud environment
💡 Automate periodic tasks using cron scheduling
Strengths & Limitations
✅ Strengths
- • Actively maintained, recent updates
- • High community interest (41.7k stars)
- • Permissive open-source license (Apache-2.0)
- • Established track record (4 years in production)
Categories
Quick Start
```bash
pip install phidata
```
```python
from phi.agent import Agent
from phi.model.openai import OpenAIChat
agent = Agent(
model=OpenAIChat(id="gpt-4o"),
description="You are a helpful assistant.",
tools=[],
)
agent.run("What is the capital of France?")
```